Vascular Catheter Site Care: The Clinical and Economic Benefits of Chlorhexidine Gluconate Compared with Povidone Iodine

Transcription

1 MAJOR ARTICLE Vascular Catheter Site Care: The Clinical and Economic Benefits of Chlorhexidine Gluconate Compared with Povidone Iodine Nathorn Chaiyakunapruk, 1,8 David L. Veenstra, 1 Benjamin A. Lipsky, 2,4 Sean D. Sullivan, 1,3 and Sanjay Saint 5,6,7 1 Pharmaceutical Outcomes Research and Policy Program, Department of Pharmacy, 2 School of Medicine, and 3 Department of Health Services, University of Washington, and 4 Veterans Affairs Puget Sound Health Care System, Seattle; 5 Ann Arbor Veterans Affairs Medical Center, 6 Department of Internal Medicine, University of Michigan, and 7 Patient Safety Enhancement Program, University of Michigan Health System, Ann Arbor; and 8 Department of Pharmacy Practice, Naresuan University, Pitsanuloak, Thailand The use of chlorhexidine gluconate solution for vascular catheter insertion site care reduces the risk of catheterrelated bloodstream infection by one-half, compared with povidone iodine. Our objective was to evaluate the cost-effectiveness of chlorhexidine gluconate versus povidone iodine. We used data from randomized, controlled trials, meta-analyses, and epidemiologic studies to construct a decision analysis model. We estimated that use of chlorhexidine, rather than povidone, for central catheter site care resulted in a 1.6% decrease in the incidence of catheter-related bloodstream infection, a 0.23% decrease in the incidence of death, and savings of $113 per catheter used. For peripheral catheter site care, the results were similar, although the differences were smaller. The results were found to be robust on multivariate sensitivity analyses. Use of chlorhexidine gluconate in place of the current standard solution for vascular catheter site care is a simple and cost-effective method of improving patient safety in the hospital setting. More than 150 million intravascular devices are used annually in the United States, including 15 million central vascular catheters. However, intravascular catheters can lead to infectious complications [1], such as catheter-related bloodstream infection, which has been associated with an attributable mortality of 10% 20% and increased medical care costs of approximately $1000 $15,000 [2 6]. Disinfecting the skin at the cath- Received 7 February 2003; accepted 6 May 2003; electronically published 27 August Financial support: Research Foundation for the Prevention of Complications Associated with Health Care; Pharmaceutical Research and Manufacturers of America Foundation Career Development Award in Pharmacoeconomics (to D.L.V.); Health Services Research and Development Program, Department of Veterans Affairs (Career Development Award to S.S.); Agency for Healthcare Research and Quality (Patient Safety Developmental Center grant P20-HS11540 to S.S.). Reprints or correspondence: Dr. David L. Veenstra, Pharmaceutical Outcomes Research and Policy Program, Dept. of Pharmacy, Box , University of Washington, Seattle, WA (veenstra@u.washington.edu). Clinical Infectious Diseases 2003; 37: by the Infectious Diseases Society of America. All rights reserved /2003/ $15.00 eter insertion site with antiseptic solution helps to prevent these infections, and povidone iodine is the agent most commonly used for this purpose [7 9]. A recent meta-analysis of 8 randomized, controlled trials [10 17] found that the incidence of catheter-related bloodstream infection was significantly lower among patients whose catheter sites were disinfected with chlorhexidine gluconate than among those for whom povidone iodine was used [18]. In addition, Infectious Diseases Society of America guidelines recommended the use of chlorhexidine gluconate as an antiseptic for the prevention of catheter-related infections [19]. It is unclear, however, whether the benefit of chlorhexidine gluconate is worth the extra cost. The extra cost of chlorhexidine gluconate site disinfection is relatively modest, but purchasers often balk at new interventions that are associated with even a small increase in cost, especially interventions that are likely to be widely used. In addition, whether only patients with central lines should receive chlorhexidine gluconate (or also patients with peripheral lines) and whether chlor- 764 CID 2003:37 (15 September) Chaiyakunapruk et al.

2 hexidine gluconate should replace povidone iodine throughout the hospital setting remain uncertain. Therefore, we evaluated the cost-effectiveness of the use of these 2 solutions for vascular catheter site care to clarify for decision-makers the clinical and economic consequences of using each of these solutions. METHODS Decision analysis model. A decision analysis model was developed to evaluate the outcomes associated with the use of chlorhexidine gluconate versus povidone iodine solutions for catheter site care (figure 1) [20]. Either solution could be used at the time that an intravascular catheter was inserted and then every h to cleanse the insertion site. Patients with a catheter could have 1 of 4 outcomes: (1) catheter-related bloodstream infection, defined as isolation of the identical pathogen from a peripherally obtained blood culture and from a colonized catheter; (2) local catheter-related infection, defined as the presence of purulence or signs of inflammation (e.g., erythema, tenderness, and induration) within 2 cm of the catheter exit site [21]; (3) catheter colonization without bloodstream infection or local catheter-related infection; and (4) no colonization or infectious complications. Colonization of the catheter was defined as growth of microorganisms from a catheter segment using quantitative (11000 cfu/ml) [22] or semiquantitative (115 cfu) [23] culture techniques. Catheter-related bloodstream infection was associated with a risk of dying. We assumed that local or systemic catheter-related infections did not occur without preceding catheter colonization. We performed the analysis using the hospital perspective; the time horizon was the period of hospitalization. The hypothetical cohort in the decision analysis model included hospitalized patients requiring either a peripheral or central vascular catheter for short-term use (average duration,!10 days). Because the risk of catheter-related infection differs for central and peripheral venous catheters [21], we analyzed these cohorts separately. We considered central vascular catheters to include central venous, peripherally inserted central venous, pulmonary arterial, and hemodialysis catheters and introducer sheaths, whereas peripheral vascular catheters included peripheral venous and peripheral arterial catheters. Likelihood of events. The probabilities of clinical events used in the decision analysis model are shown in table 1. The probabilities of catheter-related bloodstream infection and catheter colonization with povidone iodine (the baseline risk) were derived by statistically pooling data from 8 randomized, controlled trials that compared chlorhexidine gluconate with povidone iodine solution for insertion site care [10 17]. The probability of catheter-related bloodstream infection with chlorhexidine gluconate was determined by multiplying the probability of catheter-related bloodstream infection when povidone iodine was used by the summary risk ratio of catheter- Figure 1. Decision tree comparing the use of chlorhexidine gluconate with the use of povidone iodine for vascular catheter site care Chlorhexidine vs. Povidone CID 2003:37 (15 September) 765

3 Table 1. Probabilities of clinical events and comparison of the costs associated with chlorhexidine gluconate and povidone iodine solutions for vascular catheter site care. Probability and risk ratio of clinical event or cost Base-case value (range) Central line Peripheral line Reference(s) Catheter-related bloodstream infection Probability when povidone iodine solution is used, % 3.08 ( ) 0.92 ( ) [10 17] Risk ratio for chlorhexidine gluconate solution a 0.49 ( ) 0.49 ( ) [10 17] Catheter colonization Probability when povidone iodine solution is used, % ( ) 7.91 ( ) [10 17] Risk ratio for chlorhexidine gluconate solution a 0.49 ( ) 0.49 ( ) [10 17] Probability of death attributable to catheter-related bloodstream infection, % 15.0 ( ) 1.12 ( ) [5, 24 29] Probability of local infection if colonization occurs, % 20.0 ( ) 20.0 ( ) [30] Cost in the United States in 2001, $ Additional cost of chlorhexidine gluconate compared with povidone iodine 1.53 ( ) 0.12 ( ) See Methods Cost associated with catheter-related bloodstream infection 7113 (812 12,395) 1534 ( ) See Methods Cost of managing local infection 165 ( ) 144 (89 322) See Methods a To introduce a correlation between the probabilities of events in each treatment arm, the probability associated with the use of a chlorhexidine gluconate solution was calculated by multiplying the risk ratio by the probability associated with the use of a povidone iodine solution. In the base case for the chlorhexidine arm, the probabilities that a catheter-related bloodstream infection would occur were 1.51% and 0.45% for central and peripheral lines, respectively, and the probabilities of catheter colonization were 8.86% and 3.88%. related bloodstream infection when chlorhexidine gluconate was used (table 1), based on the results of the recently published meta-analysis [18]. The probability of catheter colonization when chlorhexidine gluconate was used was similarly derived. We estimated that 20% of the colonized catheters were associated with local signs of infection [30]. All probabilities were calculated separately for central and peripheral catheter models. The probability of death attributable to catheter-related bloodstream infection for a central venous catheter was calculated on the basis of data from studies published elsewhere, which report excess mortality of 4% 25% [22, 24 29]. We used a 15% attributable mortality for the base-case scenario and explored a range from 4% to 25% in sensitivity analyses. For patients with peripheral vascular catheters, we assumed the same relative risk of death due to catheter-related bloodstream infection and a baseline risk of death of 2.6% (range, 1.0% 5.0%) [31] and estimated an attributable mortality of 1.1% (range, 0.5% 2.1%). Costs. We estimated the cost of catheter-related bloodstream infection and local catheter-related infection using a microcosting technique. The estimate of the increase in the length of hospital stay due to catheter-related bloodstream infection was based on current literature [5, 25 27, 29]. Medical use costs for patients with catheter-related local or bloodstream infection were estimated from published sources, supplemented with expert opinion when necessary. We used the unit cost of each hospital day in the intensive care unit and on a regular hospital ward, of laboratory fees, and of medications from the University of Washington Medical Center (Seattle). We based our data on professional fees on Health Care Financing Administration fee schedules for the year The total cost of catheter-related bloodstream infection in a patient with a central catheter included the cost of 1 catheter tip culture and 3 blood cultures and antibiotic susceptibility tests (range, 2 5 cultures and tests), 7 days of intravenous vancomycin therapy (range, 2 14 days), no days of oral dicloxacillin therapy (range, 0 5 days), 5 extra days in an intensive care unit (range, 0 10 days), 2 extra days on a general hospital ward (range, 1 4 days), and professional fees [32]. For a peripheral catheter, the cost of catheter-related bloodstream infection was similar, but we estimated that patients would require only 2 extra days in general hospital units (range, 0 2 days). Using these assumptions, we estimated that the cost of a catheter-related bloodstream infection would be $7113 (range, $812 $12,395) for a central catheter and $1534 (range, $155 $1803) for a peripheral catheter. We estimated that treatment of local catheter-related infection would require 7 days of intravenous vancomycin therapy (range, 2 14 days), no days of oral dicloxacillin therapy (range, 0 5 days), 1 blood culture (range, 1 3 cultures), 1 catheter tip culture, and 1 catheter replacement. The total cost estimated for managing local catheter-related infections was $165 (range, $109 $343) for central catheters, and $144 (range, $89 $322) for peripheral catheters. We estimated the incremental cost of chlorhexidine glucon- 766 CID 2003:37 (15 September) Chaiyakunapruk et al.

4 Table 2. Results of decision analysis comparing chlorhexidine gluconate and povidone iodine solutions for vascular catheter site care in the United States in Catheter type, variable Direct medical costs, $ Incidence of CR-BSI, % Incidence of death due to CR-BSI, % Central line Chlorhexidine gluconate solution Povidone iodine solution Difference (95% central range) a 113 (18 241) 1.6 ( ) 0.23 ( ) Best-case scenario Worst-case scenario Peripheral line Chlorhexidine gluconate solution Povidone iodine solution Difference (95% central range) a 8 (2 20) 0.5 ( ) ( ) Best-case scenario Worst-case scenario NOTE. Minus signs indicate a decrease in cost or in the incidence of CR-BSI infection or death due to CR- BSI. CR-BSI, catheter-related bloodstream infection. a Ninety-five percent central range from multivariate sensitivity analysis. ate over povidone iodine under a variety of clinical situations. Different packages (swabstick, pad, and bottle) and amounts (3 30 ml) of povidone iodine solution are used for catheter site care, and alcohol solution is sometimes applied before the catheter site is disinfected with povidone iodine. We incorporated the diversity of clinical practices into our cost estimates to increase the generalizability of our results. We obtained the cost of products from the University of Washington Medical Center, the government federal supply schedule year 2001, and manufacturers. The total additional cost of chlorhexidine gluconate, in comparison with that of povidone iodine, for site care over a duration of catheterization of 7 days (range, 2 10 days) was $1.53 (range, $0.14 $5.90) for a central catheter and $0.12 (range, $0.06 $0.18) for a peripheral catheter. Outcome assessment and sensitivity analyses. Outcomes calculated were the incidence of catheter-related bloodstream infection, the incidence of death attributable to catheter-related bloodstream infection, and direct medical costs. To assess the overall uncertainty associated with the results, we used Monte Carlo simulation to conduct a multivariate sensitivity analysis. The distributions for the parameters were fit so that the means were similar to the base case and so that the 95% central ranges corresponded with the ranges in table 1. The risk of developing clinical events in the povidone iodine arm and the risk ratios used in the simulation were assumed to follow a logistic distribution and a log-normal distribution, respectively. The cost estimates followed a normal or a log-normal distribution [33 35]. The decision analysis model was simulated 10,000 times. The mean expected value and 95% central range for the differences in the incidence of death, the incidence of catheterrelated bloodstream infection, and direct medical costs were determined. We also conducted a series of 1-way sensitivity analyses to evaluate the effect of varying individual parameters on the outcomes. To further test the robustness of the results, we set all parameters in the model to favor chlorhexidine gluconate in a best-case scenario and to favor povidone iodine in a worst-case scenario. Main assumptions in the analysis. There were several main assumptions in our analysis: (1) The relative risk of death due to catheter-related bloodstream infection was the same for central and peripheral vascular catheters. (2) The relative risks for catheter-related bloodstream infection and catheter colonization for chlorhexidine gluconate, compared with povidone iodine, were the same in central and peripheral vascular catheters. (3) The cost of catheter-related bloodstream infection was independent of survival outcome. (4) Catheter colonization without local infection had no costs or adverse outcomes. (5) Catheter-site erythema without evidence of local infection did not affect survival outcome or cost. RESULTS Costs and outcomes. In the base-case analysis, use of chlorhexidine gluconate rather than povidone iodine for central line catheter site care led to an absolute decrease in the incidence of catheter-related bloodstream infection of 16 cases/1000 catheters (from 31 to 15 cases/1000 catheters; number needed to treat, 64 patients), and a decrease in the incidence of death attributable to catheter-related bloodstream infection of 2 cases/ 1000 catheters (from 4 to 2 cases/1000 catheters; number Chlorhexidine vs. Povidone CID 2003:37 (15 September) 767

5 needed to treat, 435 patients) (table 2). In addition to these clinical benefits, use of chlorhexidine gluconate resulted in expected cost savings of $113 for each catheter used, compared with use of povidone iodine. Use of chlorhexidine gluconate rather than povidone iodine for peripheral line catheter site care led to an absolute decrease in the incidence of catheter-related bloodstream infection of 5 cases/1000 catheters (from 9 to 4 cases/1000 catheters; number needed to treat, 213 patients) and a decrease in the incidence of death attributable to catheter-related bloodstream infection of 0.05 cases/1000 catheters (from 0.10 to 0.05 cases/1000 catheters; number needed to treat, 21,277 patients). In addition to these clinical benefits, use of chlorhexidine gluconate resulted in expected cost savings of $8 per catheter, compared with use of povidone iodine. Sensitivity analyses. As shown in figure 2, use of chlorhexidine gluconate for central catheter site care remained the best strategy in all of the 1-way sensitivity analyses. The cost of catheter-related bloodstream infection was the most influential parameter in the model. Other influential parameters included the reduction in risk of catheter-related bloodstream infection for chlorhexidine gluconate, the probability of death due to catheter-related bloodstream infection, and the baseline risk of catheter-related bloodstream infection. In calculations based on the 95% central range of values obtained from the probabilistic sensitivity analysis, use of chlorhexidine gluconate rather than povidone iodine for central vascular catheter site care was found to reduce the incidence of death by 1 5 cases/1000 catheters and of catheter-related bloodstream infection by 6 25 cases/1000 catheters. The cost savings ranged from $18 to $241 per catheter used. In the Monte Carlo analysis, use of chlorhexidine gluconate resulted in cost savings in 199% of the simulations (figure 3). Use of chlorhexidine gluconate resulted in a dominant strategy in the best-case scenario but not in the worst-case scenario; in the worst-case scenario, use of chlorhexidine gluconate resulted in an increase of total medical cost of $4. In this scenario, however, the use of chlorhexidine gluconate still resulted in a decrease in the incidence of catheter-related bloodstream infection (decrease of 0.23%) and death (decrease of 0.01%). For peripheral vascular catheters, use of chlorhexidine gluconate for insertion site care was again found to be the best strategy in all 1-way sensitivity analyses. The baseline risk of catheter-related bloodstream infection was the most influential parameter in the model. When the base-case scenario parameters were used in calculations, the use of chlorhexidine gluconate would save costs as long as the total additional cost of chlorhexidine gluconate was!$8.20. In the probabilistic sensitivity analysis, use of chlorhexidine gluconate was found to reduce the incidence of death by cases/1000 catheters and the incidence of catheter-related bloodstream infection by 1 12 cases/1000 catheters. In the Monte Carlo analysis, use of chlorhexidine gluconate produced cost savings in 199% of the simulations. In the worst-case scenario, use of chlorhexidine gluconate resulted in an increase in direct medical costs of $0.18 per catheter but did not affect the incidence of catheter-related Figure 2. Sensitivity analyses of central vascular catheter site care: total costs associated with use of chlorhexidine gluconate, compared with those for use of povidone iodine. Threshold values represent parameter values that result in no difference in cost between 2 antiseptic solutions. Minus signs indicate cost savings. CR-BSI, catheter-related bloodstream infection. 768 CID 2003:37 (15 September) Chaiyakunapruk et al.

6 Figure 3. Probabilistic sensitivity analysis of the use of chlorhexidine gluconate, compared with povidone iodine, for central vascular catheter site care. All chances of the results of use of chlorhexidine gluconate for vascular catheter site care, compared with povidone iodine, are shown. Each dot, generated from a simulation using probabilistic sensitivity analysis technique, represents a chance of the outcome falling into 1 of the quadrants. bloodstream infection and death. The cost savings expected with peripheral catheters are slightly more certain than those expected with central catheters, because a much narrower distribution was used for the cost of peripheral catheter related bloodstream infection. DISCUSSION Our analysis suggests that use of chlorhexidine gluconate for patients requiring short-term vascular catheterization, either with central or peripheral catheters, will reduce the incidence of vascular catheter related infections and decrease health care costs. These results held true over a wide range of clinical and economic assumptions. This unusual combination of clinical benefits and decreased costs makes chlorhexidine gluconate attractive for routine use for both central and peripheral vascular catheter site skin care. Some of the parameters in our analysis are uncertain. Data on the probability of death due to catheter-related bloodstream infection were conflicting [24]. Although some studies have not found a significant increase in risk of death among patients with catheter-related bloodstream infection [26, 27], overall, the results of studies that examined this association are consistent with an attributable mortality of 10% 20% [28, 29]. The cost of catheter-related bloodstream infection was derived using estimates of medical use. To improve the robustness of our analysis, we used a wide range of excess hospital days, the number of laboratory orders, and a variety of drug use patterns. The baseline risk values that we used for central and peripheral catheter-related bloodstream infection are similar to the rates reported elsewhere [36 39]. Because these varied up to 3.5- fold, we used a range of values in our analysis. Different types of chlorhexidine gluconate solution were used in the individual trials that we used to populate our model, including 0.5% or 1% chlorhexidine gluconate alcoholic solution and 0.5% or 2% chlorhexidine gluconate aqueous solution. In our previously published meta-analysis, subset analyses of aqueous and nonaqueous solutions showed similar effect sizes, but only the subset analysis of the 5 studies in which alcoholic solution was used produced a statistically significant reduction in catheter-related bloodstream infections [18]. We concluded that the failure of chlorhexidine gluconate aqueous solution to achieve a significant difference, compared with povidone iodine, may have been due to inadequate statistical power. Despite these shortcomings, all of these solutions provide a concentration of chlorhexidine gluconate that is higher than the MIC for most nosocomial bacteria and yeasts [40]. Several cases of hypersensitivity reactions associated with Chlorhexidine vs. Povidone CID 2003:37 (15 September) 769

7 chlorhexidine gluconate coated vascular catheters have been reported [41], but we found none in patients for whom chlorhexidine gluconate was used on intact skin. Including hypersensitivity reactions in the model had a minimal effect on the results. Antibiotic resistance is also a concern when a choice among different antiseptic solutions is being made. Because reports of primary or acquired nosocomial resistance to chlorhexidine gluconate are rare, despite widespread use over decades [40, 42, 43], the effect on antimicrobial resistance is likely to be small. There may be other consequences of use of chlorhexidine gluconate solution. Because it is associated with a lower risk of catheter colonization and local catheter-related infection than is povidone iodine, it may decrease the incidence of unnecessary removal of catheters due to suspected infection. In contrast, the increased risk of skin erythema associated with chlorhexidine gluconate may result in the unnecessary removal of a catheter due to suspected infection [44]. Our analysis was conducted from the perspective of the health care provider, rather than for society as a whole, as is recommended by some guidelines [45]. However, from a societal perspective, including indirect costs, such as time lost from work, in the analysis would result in even greater cost savings for the chlorhexidine gluconate strategy. The efficacy data we used were obtained from a meta-analysis of clinical trials that included only patients requiring shortterm catheterization (average duration,!10 days). Therefore, the policy implications should be limited to such groups of patients. The results are otherwise likely generalizable to a variety of clinical settings. The clinical trials included in the metaanalysis were conducted in a variety of patient populations in different types of medical centers. Importantly, the magnitude of risk reduction for catheter-related bloodstream infection was consistent across studies, regardless of the type of catheter used or the setting in which care was provided. What would be the role of chlorhexidine gluconate in preventing catheter-related bloodstream infection, compared with other preventive measures? Antimicrobial-coated vascular catheters have been shown to decrease catheter-related bloodstream infection to a degree similar to that of use of chlorhexidine gluconate for catheter site care [20, 46], but the additional cost is greater (approximately $20 $40 per catheter). However, these technologies have not been directly compared or studied in combination. None of the studies in the meta-analysis of chlorhexidine gluconate solution used antimicrobial-coated catheters, and, in a separate meta-analysis of chlorhexidine-coated catheters, only 2 of 13 studies used chlorhexidine gluconate for catheter site care, which makes an evaluation of chlorhexidine gluconate versus povidone iodine for catheter site care in conjunction with the use of antimicrobial catheters unfeasible. Full barrier precautions during catheter insertion also likely reduce catheter-related bloodstream infection [47], but adherence to such procedures is poor, and educational interventions are time-consuming and expensive [48]. Only 1 of the studies included in the meta-analysis stated that maximal barrier precautions were used [16]. Pending an evaluation of the comparative effectiveness of these interventions, we believe that the use of chlorhexidine gluconate instead of povidone iodine is a clinically sound, economically attractive, and easily instituted intervention. Our analysis suggests that use of chlorhexidine gluconate rather than povidone iodine for vascular catheter site disinfection in hospitalized patients requiring short-term vascular access is likely to result in decreased morbidity, mortality, and health care costs. This simple method can be relatively easily implemented to improve patient safety, and, thus, should perhaps take priority in efforts to prevent vascular catheter related infection. References 1. Raad I. Intravascular-catheter related infections. Lancet 1998; 351: Smith RL, Meixler SM, Simberkoff MS. Excess mortality in critically ill patients with nosocomial bloodstream infections. Chest 1991; 100: Martin MA, Pfaller MA, Wenzel RP. Coagulase-negative staphylococcal bacteremia: mortality and hospital stay. Ann Intern Med 1989; 110: Haley RW, Schaberg DR, Von Allmen SD, McGowan JE Jr. Estimating the extra charges and prolongation of hospitalization due to nosocomial infections: a comparison of methods. J Infect Dis 1980; 141: Pittet D, Tarara D, Wenzel RP. Nosocomial bloodstream infection in critically ill patients: excess length of stay, extra costs, and attributable mortality. JAMA 1994; 271: Arnow PM, Quimosing EM, Beach M. Consequences of intravascular catheter sepsis. Clin Infect Dis 1993; 16: Clemence MA, Walker D, Farr BM. Central venous catheter practices: results of a survey. Am J Infect Control 1995; 23: Sellors JE, Cyna AM, Simmons SW. Aseptic precautions for inserting an epidural catheter: a survey of obstetric anaesthetists. Anaesthesia 2002; 57: Kumwenda MJ, Wright FK, Haybittle KJ. Survey of permanent central venous catheters for haemodialysis in the UK. Nephrol Dial Transplant 1996; 11: Maki DG, Ringer M, Alvarado CJ. Prospective randomised trial of povidone iodine, alcohol, and chlorhexidine for prevention of infection associated with central venous and arterial catheters. Lancet 1991; 338: Mimoz O, Pieroni L, Lawrence C, et al. Prospective, randomized trial of two antiseptic solutions for prevention of central venous or arterial catheter colonization and infection in intensive care unit patients. Crit Care Med 1996; 24: Legras A, Cattier B, Dequin PF, Boulain T, Perrotin D. Etude prospective randomisee pour la prevention des infections liees aux catheters: chlorhexidine alcoolique contre polyvidone iodee. Reanimation Urgences 1997; 6: Meffre C, Girard R, Hajjar J, Fabry J. Is peripheral venous catheter colonisation related to the antiseptic used for disinfection of the insertion site? Povidone iodine vs. alcoholic chlorhexidine: a multicenter randomised prospective study. Catheter Study Group. Hygienes 1995; 9: CID 2003:37 (15 September) Chaiyakunapruk et al.

8 14. LeBlanc A, Cobett S. IV site infection: a prospective, randomized clinical trial comparing the efficacy of three methods of skin antisepsis. CINA Conference 99. CINA J 1999; 15: Sheehan G, Leicht K, O Brien M, Taylor G, Rennie R. Chlorhexidine versus povidone iodine as cutaneous antisepsis for prevention of vascular-catheter infection [abstract 1616]. In: Program and abstracts of the 33rd Interscience Conference on Antimicrobial Agents and Chemotherapy (New Orleans). Washington, DC: American Society for Microbiology, 1993: Humar A, Ostromecki A, Direnfeld J, et al. Prospective randomized trial of 10% povidone iodine versus 0.5% tincture of chlorhexidine as cutaneous antisepsis for prevention of central venous catheter infection. Clin Infect Dis 2000; 31: Knasinski V, Maki DG. A prospective, randomized, controlled trial of 1% chlorhexidine 75% alcohol vs. 10% povidone iodine for cutaneous disinfection and follow-up site care with central venous and arterial catheters [abstract]. Presented at: 14th annual conference of the National Association of Vascular Access Network (San Diego), September Chaiyakunapruk N, Veenstra DL, Lipsky B, Saint S. Chlorhexidine compared with povidone-iodine solution for vascular catheter site care: a meta-analysis. Ann Intern Med 2002; 136: O Grady NP, Alexander M, Dellinger EP, et al. Guidelines for the prevention of intravascular catheter related infections. Clin Infect Dis 2002; 35: Veenstra DL, Saint S, Sullivan SD. Cost-effectiveness of antisepticimpregnated central venous catheters for the prevention of catheterrelated bloodstream infection. JAMA 1999; 282: Pearson ML. Guideline for prevention of intravascular device related infections. Part I. Intravascular device related infections: an overview. The Hospital Infection Control Practices Advisory Committee. Am J Infect Control 1996; 24: Brun-Buisson C, Abrouk F, Legrand P, Huet Y, Larabi S, Rapin M. Diagnosis of central venous catheter related sepsis: critical level of quantitative tip cultures. Arch Intern Med 1987; 147: Maki DG, Weise CE, Sarafin HW. A semiquantitative culture method for identifying intravenous-catheter related infection. N Engl J Med 1977; 296: Saint S, Veenstra DL, Lipsky BA. The clinical and economic consequences of nosocomial central venous catheter related infection: are antimicrobial catheters useful? Infect Control Hosp Epidemiol 2000; 21: Pittet D, Wenzel RP. Nosocomial bloodstream infections in the critically ill. JAMA 1994; 272: Digiovine B, Chenoweth C, Watts C, Higgins M. The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit. Am J Respir Crit Care Med 1999; 160: Rello J, Ochagavia A, Sabanes E, et al. Evaluation of outcome of intravenous catheter related infections in critically ill patients. Am J Respir Crit Care Med 2000; 162: Soufir L, Timsit JF, Mahe C, Carlet J, Regnier B, Chevret S. Attributable morbidity and mortality of catheter-related septicemia in critically ill patients: a matched, risk-adjusted, cohort study. Infect Control Hosp Epidemiol 1999; 20: Renaud B, Brun-Buisson C. Outcomes of primary and catheter-related bacteremia: a cohort and case-control study in critically ill patients. Am J Respir Crit Care Med 2001; 163: Tacconelli E, Tumbarello M, Pittiruti M, et al. Central venous catheter related sepsis in a cohort of 366 hospitalised patients. Eur J Clin Microbiol Infect Dis 1997; 16: Centers for Disease Control and Prevention. National Hospital Discharge Survey: Available at: major/hdasd/nhdsdes.htm. Accessed 11 August Digiovine B, Chenoweth C, Watts C, Higgins M. The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit. Am J Respir Crit Care Med 1999; 160: Doubilet P, Begg CB, Weinstein MC, Braun P, McNeil BJ. Probabilistic sensitivity analysis using Monte Carlo simulation: a practical approach. Med Decis Making 1985; 5: Critchfield GC, Willard KE. Probabilistic analysis of decision trees using Monte Carlo simulation. Med Decis Making 1986; 6: Merz JF, Small MJ, Fischbeck PS. Measuring decision sensitivity: a combined Monte Carlo logistic regression approach. Med Decis Making 1992; 12: National Nosocomial Infections Surveillance System, Centers for Disease Control and Prevention. Semiannual report: aggregated data from the National Nosocomial Infections Surveillance (NNIS) System: December Available at: DEC2000sar.pdf. Accessed 25 June Shimandle RB, Johnson D, Baker M, Stotland N, Karrison T, Arnow PM. Safety of peripheral intravenous catheters in children. Infect Control Hosp Epidemiol 1999; 20: Soifer NE, Borzak S, Edlin BR, Weinstein RA. Prevention of peripheral venous catheter complications with an intravenous therapy team: a randomized controlled trial. Arch Intern Med 1998; 158: Maki DG, Mermel LA. Infections due to infusion therapy. In: Bennett JV, Brachman PS, eds. Hospital infections. 4th ed. Philadelphia: Lippincott-Raven, 1998: Freney J, Husson MO, Gavini F, Madier S, Martra A, Izard D. Susceptibilities to antibiotics and antiseptics of new species of the family Enterobacteriaceae. Antimicrob Agents Chemother 1988; 32: Oda T, Hamasaki J, Kanda N, Mikami K. Anaphylactic shock induced by an antiseptic-coated central venous catheter. Anesthesiology 1997; 87: Kahan A, Philippon A, Paul G, Weber S, Richard C, Hazebroucq G. Nosocomial infections by chlorhexidine solution contaminated with Pseudomonas pickettii (Biovar VA-I). J Infect 1983; 7: Stickler DJ, Thomas B, Clayton CL, Chawla JC. Studies of the genetic basis of chlorhexidine resistance. Br J Clin Pract 1983; 25: Maki DG, Ringer M, Alvarado CJ. Prospective randomised trial of povidone iodine, alcohol, and chlorhexidine for prevention of infection associated with central venous and arterial catheters. Lancet 1991; 338: Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-effectiveness in health and medicine. New York: Oxford University Press, Veenstra DL, Saint S, Saha S, Lumley T, Sullivan SD. Efficacy of antiseptic-impregnated central venous catheters in preventing catheterrelated bloodstream infection: a meta-analysis. JAMA 1999; 281: Raad II, Hohn DC, Gilbreath BJ, et al. Prevention of central venous catheter related infections by using maximal sterile barrier precautions during insertion. Infect Control Hosp Epidemiol 1994; 15: Sherertz RJ, Wesley E, Westbrook DM, et al. Education of physiciansin-training can decrease the risk for vascular catheter infection. Ann Intern Med 2000; 132: Chlorhexidine vs. Povidone CID 2003:37 (15 September) 771